Abstract
The goal of no-reference image quality assessment (NR-IQA) is to evaluate their perceptual quality of digital images without using the distortion-free, pristine counterparts. NR-IQA is an important part of multimedia signal processing since digital images can undergo a wide variety of distortions during storage, compression, and transmission. In this paper, we propose a novel architecture that extracts deep features from the input image at multiple scales to improve the effectiveness of feature extraction for NR-IQA using convolutional neural networks. Specifically, the proposed method extracts deep activations for local patches at multiple scales and maps them onto perceptual quality scores with the help of trained Gaussian process regressors. Extensive experiments demonstrate that the introduced algorithm performs favorably against the state-of-the-art methods on three large benchmark datasets with authentic distortions (LIVE In the Wild, KonIQ-10k, and SPAQ).
Highlights
Image quality assessment has crucial importance in the acquisition, processing, analysis, and reproduction of digital images
As in the first scale, feature maps are extracted from each image patch via the pretrained convolutional neural networks (CNN), and feature vectors are complied by running the deep feature maps through global average pooling (GAP) layers
A novel architecture for no-reference image quality assessment (NR-IQA) was proposed that—inspired by the idea of spatial pyramid pooling—extracts deep features from the input image at multiple scales to improve the effectiveness of feature extraction using convolutional neural networks
Summary
Image quality assessment has crucial importance in the acquisition, processing, analysis, and reproduction of digital images. How to design an appropriate algorithm for objectively evaluating the perceptual quality of digital images is important. With the advent of large image quality assessment databases [1,2], data-driven deep learning methods have become popular in this field. With the aim of providing an accurate image quality assessment scheme, we propose an innovative deep structure based on pretrained convolutional neural networks (CNN)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have